PENGEMBANGAN MODEL SUPPORT VECTOR MACHINES (SVM) DENGAN MEMPERBANYAK DATASET UNTUK PREDIKSI BISNIS FOREX MENGGUNAKAN METODE KERNEL TRICK

• Main Authors: adi sucipto, Akhmad Khanif Zyen
• Format: Article
• Language: Indonesian
• Published: Universitas PGRI Semarang 2017-09-01
• Series: Jurnal informatika UPGRIS
• Online Access: http://journal.upgris.ac.id/index.php/JIU/article/view/1535
• ISSN: 2460-4801; 2477-6645

There are many types of investments that can be used to generate income, such as in the form of land, houses, gold, precious metals etc., there are also in the form of financial assets such as stocks, mutual funds, bonds and money markets or capital markets. One of the investments that attract enough attention today is the capital market investment. The purpose of this study is to predict and improve the accuracy of foreign exchange rates on forex business by using the Support Vector Machine model as a model for predicting and using more data sets compared with previous research that is as many as 1558 dataset. This study uses currency exchange rate data obtained from PT. Best Profit Future Cab. Surabaya is already in the form of data consisting of open, high, low, close attributes by using the current data of Euro currency exchange rate to USA Dollar with period every 1 minutes from May 12, 2016 at 09.51 until 13 May 2016 at 12:30 As much as 1689 dataset, After conducting research using Support Vector Machine model with kernel trick method to predict Forex using current data of Euro exchange rate to USA Dollar with period every 1 minutes from May 12, 2016 at 09.51 until 13 May 2016 at 12:30 as much as 1689 The dataset yielded a considerable prediction accuracy of 97.86%, with this considerable accuracy indicating that the movement of the Euro currency exchange rate to the USA Dollar on May 12 to May 13, 2016 can be predicted precisely.